Phonocardiogram (PCG)-based in-cabin heart monitoring

Abstract

A heart monitoring system includes at least two sensors embedded in a seat, such as a driver's seat in a vehicle. One of the sensors obtains a phonocardiogram (PCG) of the driver's heart in addition to noise. Another sensor is a reference sensor that obtains a noise signal, but does not include the PCG signal. Processing circuitry receives the heart signal with the noise and the reference noise signal, and performs adaptive filtering to remove the noise from the heart signal. Further analysis detects a heart rate or other heart measurements in the heart signal, and may output an alert if a heart condition is detected.

What is claimed is: 
     
       1. A system, comprising:
 a first sensor positioned in a backrest of a vehicle seat, the first sensor positioned along a first side of the backrest to capture a target signal representative of heart sounds and noise; 
 a second sensor positioned in the backrest of the vehicle seat, the second sensor positioned along a second side of the backrest to capture a reference signal that represents, at least partially, second noise and is devoid of non-attenuated heart sounds relative to the heart sounds, with the second side being opposite the first side relative to a center line of the backrest; and 
 processing circuitry configured to:
 receive the target signal; 
 receive the reference signal; 
 filter the target signal using the reference signal to generate a filtered signal, wherein filtering the target signal removes at least a portion of the noise from the target signal; and 
 determine a heart measurement based on the filtered signal. 
 
 
     
     
       2. The system of  claim 1 , wherein the heart measurement is a heart rate, and the processing circuitry is further configured to compare the heart rate to at least one threshold value to determine whether a driver in the vehicle seat is experiencing a health issue that impairs the driver's ability to drive. 
     
     
       3. The system of  claim 1 , wherein the heart measurement is heart rate variability (HRV), and the processing circuitry is further configured to determine, based at least in part based on the HRV, whether a driver in the vehicle seat is experiencing a health issue that impairs the driver's ability to drive. 
     
     
       4. The system of  claim 1 , wherein the filtering the target signal comprises performing an adaptive filtering algorithm. 
     
     
       5. The system of  claim 4 , wherein the adaptive filtering algorithm comprises a recursive least squares (RLS) filter. 
     
     
       6. The system of  claim 1 , wherein the first sensor is one of an array of sensors at different positions along the first side of the backrest. 
     
     
       7. The system of  claim 6 , wherein the processing circuitry is further configured to select the first sensor from the array of sensors, the first sensor providing the strongest heart signal of heart signals from the array of sensors. 
     
     
       8. The system of  claim 6 , wherein the processing circuitry is further configured to:
 generate a plurality of filtered signals using the reference signal, wherein a first filtered signal of the plurality of filtered signals corresponds to a first target signal from the array of sensors, and wherein a second filtered signal of the plurality of filtered signals corresponds to a second target signal from the array of sensors; and 
 determine the heart measurement based on the plurality of filtered signals. 
 
     
     
       9. The system of  claim 1 , wherein the first sensor is a first acoustic sensor, and the second sensor is a second acoustic sensor. 
     
     
       10. The system of  claim 1 , wherein the first sensor is a first piezoelectric sensor, and the second sensor is a second piezoelectric sensor. 
     
     
       11. A method, comprising:
 obtaining a target signal from a first sensor positioned in a backrest of a vehicle seat, the first sensor positioned along a first side of the backrest to capture heart sounds and noise; 
 obtaining a reference signal from a second sensor positioned in the backrest of the vehicle seat, the second sensor positioned along a second side of the backrest, with the second side opposite the first side relative to a center line of the backrest, wherein the reference signal is representative, at least partially, of second noise and is devoid of non-attenuated heart sounds relative to the heart sounds; 
 filtering, by processing circuitry, the target signal using the reference signal to generate a filtered signal, wherein filtering the target signal removes at least a portion of the noise from the target signal; and 
 determining, by the processing circuitry, a heart measurement based on the filtered signal. 
 
     
     
       12. The method of  claim 11 , wherein the heart measurement is a heart rate, the method further comprising:
 comparing the heart rate to at least one threshold value to determine that a driver in the vehicle seat is experiencing a health issue that impairs the driver's ability to drive; and 
 generating an alert in response to determining that the driver is experiencing the health issue. 
 
     
     
       13. The method of  claim 11 , wherein the heart measurement is heart rate variability (HRV), the method further comprising:
 determining based at least in part based on the HRV, that a driver in the vehicle seat is experiencing a health issue that impairs the driver's ability to drive; and 
 generating an alert in response to determining that the driver is experiencing the health issue. 
 
     
     
       14. The method of  claim 11 , wherein the filtering the target signal comprises applying a recursive least squares (RLS) filter to the target signal. 
     
     
       15. The method of  claim 11 , wherein the first sensor is one of an array of sensors at different positions along the first side of the backrest, the method further comprising selecting the first sensor from the array of sensors based on the first sensor providing the strongest heart signal of heart signals from the array of sensors. 
     
     
       16. The method of  claim 11 , wherein the first sensor is a first acoustic sensor, and the second sensor is a second acoustic sensor. 
     
     
       17. The method of  claim 11 , wherein the first sensor is a first piezoelectric sensor, and the second sensor is a second piezoelectric sensor. 
     
     
       18. A non-transitory computer-readable media having instructions stored thereon, wherein the instructions, when executed by a device, cause the device to:
 obtain a target signal from a first sensor positioned in a backrest of a vehicle seat, the first sensor positioned along a first side of the backrest to capture heart sounds and noise; 
 obtain a reference signal from a second sensor positioned in the backrest of the vehicle seat, the second sensor positioned along a second side of the backrest, with the second side being opposite the first side relative to a center line of the backrest, wherein the reference signal is representative, at least partially, of second noise and is devoid of non-attenuated heart sounds relative to the heart sounds; 
 filter the target signal using the reference signal to generate a filtered signal, wherein filtering the target signal removes at least a portion of the noise from the target signal; and 
 determine a heart measurement based on the filtered signal. 
 
     
     
       19. The non-transitory computer-readable media of  claim 18 , wherein the heart measurement is a heart rate, and the instructions, when executed by the device, further cause the device to:
 compare the heart rate to at least one threshold value to determine that a driver in the vehicle seat is experiencing a health issue that impairs the driver's ability to drive; and 
 generate an alert in response to determining that the driver is experiencing the health issue. 
 
     
     
       20. The non-transitory computer-readable media of  claim 18 , wherein filtering the target signal comprises applying a recursive least squares (RLS) filter to the target signal.